kinga-marszalkowska / mongo_dart

Mongo_dart: MongoDB driver for Dart programming language

Mongo-dart - MongoDB driver for Dart programming language

Server-side driver library for MongoDb implemented in pure Dart.

Development release

Please, note that this is a development release, published only to permit testing and suggestions from the users. Do not use in production. See OP_MSG section

Basic usage

Obtaining connection

  var db = Db("mongodb://localhost:27017/mongo_dart-blog");
  await db.open();

or

  var db = await Db.create("mongodb+srv://<user>:<password>@<host>:<port>/<database-name>?<parameters>");
  await db.open();

Querying

Method find returns stream of maps and accept query parameters, usually build by fluent API query builder provided by mongo_dart_query as top level getter where

  var coll = db.collection('user');
  await coll.find(where.lt("age", 18)).toList();
  
  //....
  
  await coll
      .find(where.gt("my_field", 995).sortBy('my_field'))
      .forEach((v) => print(v));

  //....
  
  await coll.find(where.sortBy('itemId').skip(300).limit(25)).toList();
  

Method findOne take the same parameter and returns Future of just one map (mongo document)

  val = await coll.findOne(where.eq("my_field", 17).fields(['str_field','my_field']));

Take notice in these samples that unlike mongo shell such parameters as projection (fields), limit and skip are passed as part of regular query through query builder

Inserting documents

  await usersCollection.insertAll([
    {'login': 'jdoe', 'name': 'John Doe', 'email': '[email protected]'},
    {'login': 'lsmith', 'name': 'Lucy Smith', 'email': '[email protected]'}
  ]);

Updating documents

You can update whole document with method save

  var v1 = await coll.findOne({"name": "c"});
  v1["value"] = 31;
  await coll.save(v1);

or you can perform field level updates with method update and top level getter modify for ModifierBuilder fluent API

  coll.update(where.eq('name', 'Daniel Robinson'), modify.set('age', 31));

Removing documents

  students.remove(where.id(id));
  /// or, to remove all documents from collection
  students.remove();

Simple app on base of JSON ZIPS dataset

import 'package:mongo_dart/mongo_dart.dart';

void main() async {
  void displayZip(Map zip) {
    print(
        'state: ${zip["state"]}, city: ${zip["city"]}, zip: ${zip["id"]}, population: ${zip["pop"]}');
  }
  var db =
      Db("mongodb://reader:[email protected]:37468/samlple");
  var zips = db.collection('zip');
  await db.open();
  print('''
******************** Zips for state NY, with population between 14000 and 16000,
******************** reverse ordered by population''');
  await zips
      .find(where
          .eq('state', 'NY')
          .inRange('pop', 14000, 16000)
          .sortBy('pop', descending: true))
      .forEach(displayZip);
  print('\n******************** Find ZIP for code 78829 (BATESVILLE)');
  var batesville = await zips.findOne(where.eq('id', '78829'));
  displayZip(batesville);
  print('******************** Find 10 ZIP closest to BATESVILLE');
  await zips
      .find(where.near('loc', batesville["loc"]).limit(10))
      .forEach(displayZip);
  print('closing db');
  await db.close();
}

Building aggregation queries

import 'package: mongo_dart/mongo_dart.dart';

main() async {
  final db = Db('mongodb://127.0.0.1/testdb');
  final pipeline = AggregationPipelineBuilder()
    .addStage(
      Match(where.eq('status', 'A').map['\$query']))
    .addStage(
      Group(
        id: Field('cust_id'),
        fields: {
          'total': Sum(Field('amount'))
        }
      )).build();
  final result =
    await DbCollection(db, 'orders')
      .aggregateToStream(pipeline).toList();
  result.forEach(print);
}

Secure connection

You can connect using a secured tls/ssl connection in one of these two ways:

• setting the secure connection parameter to true in db.open()

    await db.open(secure: true);

• adding a query parameter => "tls=true" (or "ssl=true").

    var db = DB('mongodb://www.example.com:27017/test?tls=true&authSource=admin');
              or
    var db = DB('mongodb://www.example.com:27017/test?ssl=true&authSource=admin');

When you use the mongodb+srv url schema, the "tls" (or "ssl") parameter is implicitly considered true.

No certificates can be used.

Atlas (MongoDb cloud service) connection

Atlas requires a tls connection, so now it is possible to connect to this cloud service. When creating a cluster Atlas shows you three ways of connecting: Mongo shell, Driver and MongoDb Compass Application. The connection string is in Seedlist Connection Format (starts with mongodb+srv://).

Take the Url related to the Driver mode and pass it to Db.create().

Please, not that Db.create is an asynchronous constructor, so you have to await for it.

  var db = await Db.create("mongodb+srv://<user>:<password>@<host>:<port>/<database>?<parameters>");
  await db.open();

OP_MSG

Only for those using mongodb version 3.6 or later State: experimental

Donate If you need a reliable driver like I do, please help me maintaining this software with a donation. This will allow me to dedicate more time on development, testing and documentation.

Starting from 3.6, a new protocol has been developed form mongodb to exchange data with the server. We are updating all commands and operation so that we can use the new protocol. This will be done gradually. At present we have developed the following operations:

Command	Status	Legacy alternative	Notes
insertOne	Experimental	insert	insertOne was already present in the previous release, but some changes in the call have been made
insertMany	Experimental	insert	Allows to insert many documents in one commnand. It is subject to the max number of documents per message limit (at present <4.4> 100,000)
deleteOne	Experimental	-	Allows to delete one document. Example Example 2
deleteMany	Experimental	remove	Allows to delete many documents in one commnand. Example
modernFind	Experimental	legacyFind	The old method find now is simply a wrapper. If you are running mongoDb 3.6 or later, modernFind is called, otherwise legacyFind
modernAggregate	Experimental	legacyAggregateToStream	The old method aggregateToStream now is simply a wrapper. If you are running mongoDb 3.6 or later, modernAggregate is called, otherwise legacyAggregateToStream
modernAggregateCursor	Experimental	-	This method has no corresponding legacy one, as it returns a cursor and not a Map like the legacy aggregate
-	Production	aggregate	The legacy method aggregate has no corresponding "moderm" method. It returns a Map. Is it used?
watch	Experimental	-	This method has no corresponding legacy methods. Allows to monitor changes to a collection
watchCursor	Experimental	-	This method has no corresponding legacy methods. Like watch, but resturns a cursor instead of a stream. Normally you will only need watch

Something on the bulk write side has been done too: There is the new collection helper method bulkWrite. The syntax is quite similar to that of the shell: collection.bulkWrite([{'insertOne': {'document': { "_id" : 1, "char" : "Brisbane", "class" : "monk", "lvl" : 4 }}}]); Like the in the shell you can define the insertOne, deleteOne and deleteMany operations. updateOne, updateMany and replaceOne must be still developed. In addition to the shell operation, you can use also an insertMany operation, with the following sybtax: collection.bulkWrite([{'insertMany': {'documents': [{ "_id" : 1, "char" : "Brisbane", "class" : "monk", "lvl" : 4 }]}}]); Unlike the insertMany helper, this bulk operation does not suffer of the max number of documents per message limitation.

Aside to these changes we have also to mention the new Decimal128 type management (that is backed by the Rational class). This is inherited from enhancements in the BSON package

Last but not least, some commands:

Command	Status	Legacy alternative	Notes
createIndex	Production	db.createIndex	This was needed because the legacy method didn't work any more starting from release 4.0
createCollection	Experimental	-	Explicitly creates a collection. Useful if you want to create a capped collection or assign a collation to the collection.
createView	Experimental	-	Creates a view
getParameterCommand	Experimental	-	Allows to get a specific server parameter
getAllParametersCommand	Experimental	-	Returns all config parameters
killCursorCommand	Experimental	MongoKillCursorMessage	Used internally to close an unexhausted cursor.
getMoreCommand	Experimental	MongoGetMoreMessage	Used internally to read a new batch of data from the server
getLastErrorCommand	Experimental	DbCommand.createGetLastErrorCommand	Used internally to return the status of the previous operation. It is no more needed with the modern operations, but, for compatibility reasons, it can be used still.
serverStatus	Experimental	serverStatus	The method was already present, but it has been improved giving the possiblity to return a class with all the values instead of a map

See also

• API Doc

• Status

• Recent change notes

• Additional examples and tests